The long-term objective of this research is to improve our basic understanding of the structure and function of neural circuits related to action selection. Defined as the task of resolving conflicts between competing behavioral alternatives, action selection has traditionally been carried out in non-human primates which are not amenable to a fleet of powerful experimental techniques including patch clamp recording and optogenetics. A promising approach to this impasse would be to investigate the neuronal basis of action selection in simpler organisms that are genetically tractable. The proposed research investigates the neuronal basis of action selection in the nematode Caeorhabditis elegans, an experimental system with a compact nervous system of only 302 neurons, an essentially complete anatomical wiring diagram, and a wide range of genetic, electrophysiological, and optogenetic techniques for linking molecules, genes, and neurons to behavior. This research focuses on a form of spatial orientation behavior known as klinotaxis in which rhythmic side-to-side movements of the head are biased in the direction of increasing concentration of a chemical attractant. The project is made possible by an innovative microfluidic device that presents the animal with a binary choice between fluid streams carrying different concentrations of chemoattractant, and a novel tracking system that allows one to image neuronal activity in single identified neurons in freely moving animals. The project will proceed in three phases: (1) Development of a quantitative description of klinotaxis behavior by measuring the spatiotemporal propagation of locomotory undulations and their modulation by chemoattractants in microfluidic devices. (2) Identification of the neuronal circuit for klinotaxis by optical recordings of neuronal activity, neuron- al ablations, and electrophysiological measurement of synaptic connectivity. (3) Validation of a mathematical model of action selection by photo-stimulation of identified chemosensory neurons to mimic chemosensory in- puts. The proposed research is likely to identify novel circuit motifs for action selection that can be used to generate hypotheses concerning the function of circuits regulating action selection in higher organisms. More than half of all human disease genes have a matching gene in C. elegans including diseases known to impair motor and cognitive action selection such as Parkinson's disease, Alzheimer's disease, and schizophrenia. The research is therefore likely to help trace causal connections from genetic differences to mental disorders.